trainer_state.json

{
  "best_metric": 0.8424,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0290/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 1.5021,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7786666666666666,
      "eval_loss": 0.7334726452827454,
      "eval_runtime": 5.165,
      "eval_samples_per_second": 726.044,
      "eval_steps_per_second": 11.423,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.33560091257095337,
      "learning_rate": 0.0001,
      "loss": 0.4926,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8136,
      "eval_loss": 0.625822126865387,
      "eval_runtime": 5.603,
      "eval_samples_per_second": 669.29,
      "eval_steps_per_second": 10.53,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.2906,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8352,
      "eval_loss": 0.5356540083885193,
      "eval_runtime": 5.5209,
      "eval_samples_per_second": 679.233,
      "eval_steps_per_second": 10.687,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.15521849691867828,
      "learning_rate": 0.0001,
      "loss": 0.1923,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8221333333333334,
      "eval_loss": 0.6207318902015686,
      "eval_runtime": 5.3632,
      "eval_samples_per_second": 699.204,
      "eval_steps_per_second": 11.001,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.04371129348874092,
      "learning_rate": 0.0001,
      "loss": 0.1362,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8424,
      "eval_loss": 0.5585336685180664,
      "eval_runtime": 5.6634,
      "eval_samples_per_second": 662.146,
      "eval_steps_per_second": 10.418,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}